Antioxidant Characterization of Native Monofloral Cuban Honeys

Aug 11, 2010 - Abera Belay, Gulelat Desse Haki, Marc Birringer, Hannelore Borck, Young-Chul Lee, Chang-Won Cho, Kyung-Tack Kim, Bikila Bayissa, Kaleab...
0 downloads 0 Views 1MB Size
J. Agric. Food Chem. 2010, 58, 9817–9824

9817

DOI:10.1021/jf1018164

Antioxidant Characterization of Native Monofloral Cuban Honeys JOSE M. ALVAREZ-SUAREZ,† ANA M. GONZA´LEZ- PARAMA´S,§ CELESTINO SANTOS-BUELGA,§ AND MAURIZIO BATTINO*,† †

Department of Biochemistry, Biology and Genetics, Faculty of Medicine, Marche Polytechnic University, Via Ranieri 65, 60100 Ancona, Italy, and §Grupo de Investigacio´n en Polifenoles (GIP-USAL), Faculty of Pharmacy, Salamanca University, Campus Miguel de Unamuno, E-37007 Salamanca, Spain

Five typical Cuban monofloral honeys were analyzed for their in vitro total antioxidant capacity (TAC), phenolic compounds, and ascorbic acid content. Identification and quantification of phenolics were carried out by HPLC-DAD-ESI/MS. Fourteen phenolic compounds could be identified (eight phenolic acids and six flavonoids), including three glycosylated derivatives. Similar contents of total phenolics were found in the different honeys, although they differed in their qualitative profiles. A significant (positive) correlation was found between the results of TAC obtained by parallel FIA-ABTS system and ORAC assay (r = 0.9565, p < 0.001). Similar correlations were also established between total phenolics and TAC, determined by either the ORAC (r = 0.9633; p e 0.001) or the TEAC assay (r = 0.9582; p e 0.001). Honeys were fractionated by solid-phase extraction into four fractions, and the relative contribution of each fraction to TAC was calculated. Phenolic compounds were significant contributors to the antioxidant capacity of the honeys, but they were not uniquely responsible for it. The antioxidant activity appeared to be a result of the combined activity of a range of compounds including phenolics and other minor components. Ascorbic acid was not detected. KEYWORDS: Phenolic acids; flavonoids; HPLC-MS; ORAC; TEAC

INTRODUCTION

Honey is produced by bees from plant nectars, plant secretions, and excretions of plant-sucking insects. Its composition is rather variable and primarily depends on the floral source, although certain external factors also play a role, such as seasonal and environmental factors and processing. It consists of a saturated solution of sugars, of which fructose (38%) and glucose (31%) are the main contributors, but it also contains a wide range of minor constituents, among them phenolic compounds (1, 2). Although studies on the basic composition of honeys started a hundred years ago, the interest in honey phenolics is relatively recent. Phenolic compounds in honey are mainly flavonoid aglycones and phenolic acid derivatives (benzoic and cinnamics acids and their respective esters) (3-8). Recently, the presence of certain amounts of some flavonoid glycosides has also been reported in some floral honeys (6). Phenolic compounds have a plant origin, and thus the phenolic composition in the honey varies depending on the vegetation of the area visited by the bee (4). With this in mind, phenolic compounds have been proposed as potential chemical markers for authenticating the geographical and botanical origin of honey. Flavonoids are the most common phenolics in floral honeys, and characteristic profiles could be expected in unifloral honeys depending on the corresponding plant source (9,10). It has also been

shown that a strong correlation exists between the antioxidant activity of honeys and their phenolic composition and especially the total phenolic content (11, 12). Thus, characterization of phenolics and other components in honey that might have antioxidant properties is essential to improve our knowledge about honey as a source of nutraceuticals and would also be an important tool to contribute to their authentication. Several studies on the phenolic composition have been carried out in European honeys, especially by Ferreres and co-workers (9, 10, 13-15). However, little information is available on the phenolic profiles of honeys from Cuban floral sources. The objective of this study was to identify and quantify these compounds in five different typical monofloral Cuban honeys and to determine their total antioxidant capacities. Furthermore, the contribution to the antioxidant activity of different fractions isolated from the honeys was also evaluated. MATERIALS AND METHODS

*Author to whom correspondence should be addressed (phone þ39 071 2204646; fax þ39 071 2204123; e-mail [email protected]).

Honeys Samples and Chemicals. Five different types of Cuban monofloral honeys were collected. The floral sources and number of samples analyzed were Christmas vine [Turbina corymbosa (L.) Raf; 18 samples], morning glory (Ipomoea triloba L., 16 samples), black mangrove (Avicennia germinans Jacq., 16 samples), linen vine [Govania polygama (Jack) Urb, 17 samples], and singing bean [Lysiloma latisiquum (L.) Benth, 16 samples]. All honey samples were certified by the National Center of Apiculture Research of Cuba Havana University, Cuba. Samples were collected and designated by the time of the year and the place from which each honey sample was taken; it was noted if those places coincided with

© 2010 American Chemical Society

Published on Web 08/11/2010

pubs.acs.org/JAFC

9818

J. Agric. Food Chem., Vol. 58, No. 17, 2010

the floral maps designed in that Center. All samples were tested by their organoleptic characteristics (flavor, scent), the usual available physicochemical tests [ashes (%), electrical conductivity (mS/cm), color (mm Pfund), pH, free acidity (mequiv/kg), humidity (%)], diastases index (U Schade), qualitative tests for authenticity, and HMF test (mg/kg) for quality according to the Official Methods of Analysis of the Association of Official Analytical Chemists (AOAC) (16). The botanical origin was confirmed by the traditional qualitative microscopic analysis and frequency determination of the classes of pollen grains in the honey samples (17). The different pollen morphologies were compared with reference slides from the Ecology and Systematic Institute (University of Havana, Cuba). Pollen analysis revealed that in all samples the percentage of typical pollen grains of the botanical specie was >51%, sufficient to classify them as monofloral, as reported in the above used methodology and accepted for these floral honeys (18). Fresh honey samples weighing 250 g were packed and sealed in amber glass bottles and stored at 4 °C in the dark until processing. The samples were kept at room temperature (25 ( 2 °C) overnight before analyses were performed. An artificial honey reflecting the main components of honey was prepared by dissolving 1.5 g of sucrose, 7.5 g of maltose, 40.5 g of fructose, and 33.5 g glucose in 17 mL of deionized water (19). This solution was included in the study to evaluate the contribution of the predominant sugars to the assayed activities. 2,20 -Azinobis(3-ethylbenzothiazolne-6-sulfonic acid) diammonium salt (ABTS), 6-hydroxy-2,5,7,8-tetramethyl-chroman-2-carboxylic acid (Trolox), fluorescein, and potassium chloride were purchased from Fluka Chemie (Buchs, Switzerland). 2,20 -Azobis(2-methylpropionamidine) dihydrochloride (AAPH), Amberlite XAD-2 resin, 3,4,5-trihydroxybenzoic acid (gallic acid), (þ)-catechin, sodium carbonate anhydrous, Folin-Ciocalteu’s phenol reagent, catalase, hydrogen peroxide (H2O2), ferrous sulfate (FeSO4), ascorbic acid, sucrose, maltose, fructose, and glucose were purchased from SigmaAldrich Chemie GmbH (Steinheim, Germany). All chemicals and solvents were of analytical grade. Determination of Total Phenolic and Flavonoid Content. The Folin-Ciocalteu method (20) was used to determine total phenolic content. Each honey sample (1 g) was diluted to 10 mL with distilled water and filtered through Minisart filter of 45 μm (PBI International, Milan, Italy). This solution (0.5 mL) was then mixed with 2.5 mL of 0.2 N Folin-Ciocalteu reagent for 5 min, and 2 mL of 0.7 M sodium carbonate (Na2CO3) was then added. After incubation in the dark at room temperature for 2 h, the absorbance of the reaction mixture was measured at 760 nm against sugar analogues using a Beckman Du 640 spectrophotometer (Beckman Instruments Inc., Fullerton, CA). Gallic acid was used as standard to produce the calibration curve (50-300 mg/L). The mean of three analyses was used, and the total phenolic content was expressed in gallic acid equivalents (mg of GAE/kg of honey). Total flavonoid content was determined using a colorimetric method as previously described (21). Briefly, 0.25 mL of honey solution (50%, w/v) in methanol or (þ)-catechin standard solution was mixed with 1.25 mL of distilled water in a test tube, followed by the addition of 75 μL of a 5% NaNO2 solution. After 6 min, 150 μL of a 10% AlCl3 3 6 H2O solution was added and allowed to stand for another 5 min before the addition of 0.5 mL of 1 M NaOH. The mixture was brought to 2.5 mL with distilled water and mixed well. The absorbance was immediately measured against the blank (the same mixture without the sample) at 510 nm using a spectrophotometer. The linearity range of a (þ)-catechin curve was used for calibration (5-50 mg/L). The total flavonoid content was calculated from the mean of three analyses and expressed as (þ)-catechin equivalents (mg of CE/kg of honey). Vitamin C Analysis. Vitamin C in honeys was analyzed by reversedphase HPLC, as previously described by our group (22) with minor modifications. Triplicate extracts were prepared by diluting 5 g of honey to 10 mL with dithiothreitol solution (4.2 mM in 0.1 M K2HPO4, pH 7.0) and mixing thoroughly. One milliliter of extract and 1 mL of 4.5% metaphosphoric acid were mixed, and 20 μL was injected into the HPLC. The HPLC system (Shimadzu Corp., Kyoto, Japan) consisted of a Waters 600 controller, a Waters 996 photodiode array (PDA) detector set at absorbances of 262 and 244 nm, and a column incubator at 30 °C. The HPLC column used was a YMC Pack Pro, 150  4.6 mm. A linear gradient was generated using 50 mM KH2PO4 (pH 4.5) (solvent A) and methanol (solvent B) starting at 100% A and decreasing to 70% A in 8 min. The flow rate was 0.8 mL/min.

Alvarez-Suarez et al. Total Antioxidant Capacity (TAC). The TAC of the honey samples was determined by the Trolox equivalent antioxidant capacity (TEAC) and the oxygen radical absorbance capacity (ORAC) assays. The TEAC assay was performed according to the method of Re et al. (23), partially improved by our group (24), which combines a flow injection analysis system (FIA-ABTS assay). This method is based on the ability of antioxidant compounds to quench the ABTS radical cation (ABTS•þ) and reduce the radical to the colorless neutral form. Honey was diluted in distilled water (50%, w/v) and filtered through Minisart filter of 45 μm (PBI International), and then 10 μL was injected into a serpentineknotted reaction coil and allowed to react with the ABTS•þ working solution pumped into the coil at a flow rate of 1.2 mL/min. The extent of decolorization of the reagent, expressed as percentage of absorbance inhibition, is then plotted as a function of concentrations of the antioxidants in the sample. The linearity range of the Trolox (0.03-2.5 mM) calibration curve was used. TEAC results were expressed as micromoles of Trolox equivalents per gram of honey (μmol of TE/g of honey). The mean of five analyses was used, and the results reported are as mean ( standard deviation (SD). The ORAC assay was based on the procedure previously described (25). Free radicals are produced by the radical generator AAPH, which oxidizes the fluorescent compound fluorescein, leading to a loss in fluorescence. All reagents were prepared in phosphate buffer (pH 7.0, 75 mM), and Trolox (6.25-200 μM) was used as standard. The honey samples were suitably diluted in the phosphate buffer. Each well of a 96-well microplate contained, in a final volume of 200 μL of assay solution, 150 μL of fluorescein (0.08 μM) and 25 μL of honey solution (1 mg/mL final concentration) preincubated for 10 min at 37 °C, and then 25 μL of AAPH (150 mM) was added. After the addition of AAPH, the plate was shaken automatically for 3 s, and the fluorescence was measured every 2 min for 120 min with emission and excitation wavelengths of 485 and 530 nm, respectively, using a microplate fluorescence reader (Synergy Multi-Detection Microplate Reader; Bio-Tek, Instruments, Inc., Winooski, VT) that was maintained at 37 °C. The ORAC values were calculated as area under the curve (AUC) and expressed as micromoles of Trolox equivalents per gram of honey (μmol of TE/g of honey). Fractionation of Honey on Amberlite XAD-2 Resin. The procedure for the fractionation of honey was adapted from those of Ferreres et al. (3) and Andrade et al. (14). The different honey samples (50 g each) were mixed with 5 parts of acidified water (pH 2 with HCl) until completely fluid and filtered through cotton to remove solid particles. The filtrate was then passed through a column (25  2 cm) of Amberlite XAD-2 resin (pore size = 9 nm, particle size = 0.3-1.2 mm). Sugars and polar compounds were eluted with acidified water (350 mL) (fraction 1); the column was washed with 300 mL of neutral water (fraction 2), and phenolic compounds were further eluted with methanol (600 mL). The methanol phase was concentrated under vacuum at 40 °C (B€uchi R-114, Donau, Flawil, Switzerland), suspended in water (5 mL) and extracted three times with diethyl ether (5 mL each). The ether layers were collected, evaporated under vacuum, and redissolved in methanol (fraction 3). Hesperetin was used as internal standard. The fractions were concentrated under vacuum and stored at -20 °C until further analysis. Each honey sample was fractionated and analyzed in triplicate. All of the fractions, as well as the remaining water layer after ether extraction (fraction 4), were analyzed for antioxidant activity by the ORAC assay to determine their relative contribution to the total ORAC activity of the honey. Prior to the ORAC assay, all fractions were redissolved in 5 mL of the same solvent used for their elution. When methanol or acidified water was used for fractionation, methanol or acidified water was also used in the blank and standard. Fraction 3, which was expected to contain phenolic compounds, was also analyzed by HPLC-MS. With this aim the concentrated ether layers were redissolved in methanol/H2O (50:50, v/v) to prevent the elution of the compounds in the elution front. HPLC-DAD-ESI-MS/MS Analysis of Honey Phenolics. Polyphenol identification analyses were carried out using a Hewlett-Packard 1100 series liquid chromatograph (Agilent Technologies, Waldbronn, Germany) with a quaternary pump and a diode array detector (DAD) coupled to an HP ChemStation (rev. A.05.04) data-processing station. The column used was a C18 LiChroCART (Merck, Darmstadt, Germany) (RP-18e, 250 mm  4 mm; 5 μm), operated at 35 °C. The mobile phase

Article

J. Agric. Food Chem., Vol. 58, No. 17, 2010

9819

a

Table 1. Total Antioxidant Activity of Honey and Content of Different Potential Antioxidant Components total antioxidant capacity floral source

ORAC (μmol of TE/g)

ABTS (μmol of TE/g)

phenol content (mg of GAE/kg of honey)

flavonoid content (mg of CE/kg of honey)

ascorbic acid (μg/100 g of honey)

linen vine (n = 17) morning glory (n = 16) singing bean (n = 16) black mangrove (n = 16) Christmas vine (n = 18) artificial honey

12.89 ( 0.28 a 9.26 ( 0.46 b 8.12 ( 0.23 c 7.45 ( 0.37 d 4.59 ( 0.51 e 1.09 ( 0.10 f

2.94 ( 0.23 a 2.01 ( 0.21 b 1.95 ( 0.14 b 1.22 ( 0.24 c 1.03 ( 0.28 d 0.21 ( 0.01 e

595.8 ( 16.82 a 347.5 ( 23.85 b 298.6 ( 25.01 c 233.6 ( 15.58 d 213.9 ( 14.50 d ndc

25.2 ( 0.32 a 15.8 ( 0.30 b 14.2 ( 0.27 b 17.8 ( 0.50 c 10.9 ( 0.38 d nd

udb ud ud ud ud nd

a Values are expressed as means ( standard deviation (SD). Mean values within a column sharing the same letter are not significantly different by Tukey’s multiple range test (p < 0.05). Each sample was analyzed in triplicate. b ud, undetectable. c nd, nondetermined.

consisted of H2O/formic acid (99:1, v/v) (eluent A) and methanol/ isopropanol (90:10, v/v) (eluent B). The gradient program was as follows: from 10 to 30% B over 20 min, from 30 to 40% over 10 min, from 40 to 60% B over 10 min, from 60 to 80% over 5 min and then isocratic by 5 min. The injection volume for all samples was 100 μL, and the flow rate was 1 mL/min. Identification of honey phenolics was carried out by comparing retention time and spectral characteristics of unknown analytes with standards using the HP ChemStation software (HP Hewlett-Packard ChemStation, rev. A.05.04). Spectroscopic data from all peaks were accumulated in the range of 240-400 nm. Chromatograms for the phenolic acids were recorded at 290 nm and for flavonoids at 360 nm. For calibration appropriate volumes of standard stock solutions (1000 mg/L) were diluted, and different concentration levels were analyzed. Individual phenolic acids were quantified using a calibration curve of the corresponding standard compound, and flavonoids were quantified using a quercetin calibration curve and expressed in terms of quercetin equivalents. The mass spectrometer was a Finnigan LCQ (San Jose, CA) equipped with an electrospray ionization (ESI) system and an ion trap mass analyzer, which were controlled by LCQ Xcalibur software. Nitrogen was used as both auxiliary and sheath gas at flow rates of 6 and 1.2 L/min, respectively. The capillary voltage was 10 kV and the capillary temperature, 225 °C. MS spectra were acquired in the negative and positive ionization modes between m/z 100 and 800. The MS detector was programmed to perform a series of two consecutive scans: a full scan and an MS/MS scan of the most abundant ion in the first scan, using normalized collision energy of 50%. Statistical Analysis. Statistical analysis was performed using Statistica software (Statsoft Inc., Tulsa, OK). Data were subjected to a one-way variance analysis for mean comparison, and significant differences between honey type, total ORAC, and the sum of the four fractions after the elution from the Amberlite XAD-2 resin were calculated according to HSD Tukey’s multiple-range test. Data were expressed as mean ( standard deviation (SD). Correlations were calculated on a honey mean basis, according to Pearson’s test. RESULTS AND DISCUSSION

Total Phenolic (TP) and Flavonoid (TF) Contents. The mean values and SDs of the TP content are shown in Table 1. To compare the values obtained for the five honey groups, a Tukey HSD test for comparison of means was carried out, and it was observed that the evaluated parameters showed a high power to discriminate between the different groups. According to these results, linen vine honey had the highest TP content (595.8 mg of GAE/kg of honey), followed by morning glory honey (347.5 mg of GAE/kg of honey), whereas the lowest contents were measured in black mangrove and Christmas vine (233.6 and 213.9 mg of GAE/kg of honey, respectively). TF contents are also shown in Table 1. The values varied from 10.9 mg of CE/kg in Christmas vine honey to 25.2 mg of CE/kg in linen vine honey. A significant correlation (r = 0.8697, p e 0.001) was found between TP and TF contents. The TP average values found the analyzed samples are similar to those previously reported in honeys from other origins (ranging

between 226.16 and 406.23 mg of GAE/100 g) (5, 11, 26-29). Similarly, the results obtained for TF (10-25 mg of CE/kg) are in the same range of values as previously reported for other monofloral honeys, such as eucalyptus honey (20-25 mg of CE/kg), sunflower and rape honeys (15-20 mg of CE/kg), fir, lavender, ivy, and acacia honeys (5-10 mg of CE/kg), and arbutus and chestnut honeys (